Unfortunately we haven't been running for a year yet, but, theoretically I
will be applying the incremental restores -- at least that's is how I'm
given to understand the process from our resident TSM guy. I do wish TSM
had something similar to FDR Upstream's incremental merge process so I
could have a "current" copy of an image built from incrementals on one or
two tapes.
The closest I think I've found in TSM for making a current "Hot" image copy
is the BackupSet. I've still yet to test it out, but even if it does
work, I don't think it will work like a baseline for subsequent
incrementals.
Matt Lashley
Idaho State Controller's Office
"Gowans, Chuck"
<chuck.gowans@nitck To: [EMAIL PROTECTED]
c.usda.gov> cc:
Sent by: Linux on Subject: Re: OS/390, Linux and
Disaster Recovery
390 Port
<[EMAIL PROTECTED]
T.EDU>
12/17/2002 09:38 AM
Please respond to
Linux on 390 Port
>> 3) Once the Tivoli server is up (OS/390) I restore the image to the
point
>> of disaster via CTC.
If the image was created a year ago, are you applying incremental restores
that cover the entire time span, or do you periodically snap another
baseline full-volume backup? If so, do you shutdown the guest, or do you
have some reliable method of backing it up "Hot"?
Chuck Gowans
USDA - Nat'l IT Center - Kansas City
-----Original Message-----
From: Matt Lashley [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 17, 2002 10:27 AM
To: [EMAIL PROTECTED]
Subject: Re: OS/390, Linux and Disaster Recovery
We run VM/Linux on an IFL and Tivoli under OS/390 here. Coming from an
OS/390 DR background I've got to say recovery is relatively straight
forward.
1) I make IPLable VM tapes (RES packs and such) on a weekly basis, so
bringing up the base system is a snap. There is also a tape with an
IPLable version of ICKDSF on it to use for formatting.
2) Before turning over an image to a customer I always make an initial DDR
backup of the image minidisks so that I have a bootable, network attachable
Linux guest available.
3) I also have minidisk backups of the Linux routers and internal DNS
server we use for our G-LAN based network. They're small and fairly static
so we have an internal network up very quickly.
3) Once theTivoli server is up (OS/390) I restore the image to the point
of disaster via CTC.
Then the only other piece left is for the LAN team to configure the PIX
that sits in front of the OSA attached to the Linux maser router and
connect us to the outside world. (At least until I convince my boss that a
Linux guest machine can handle the NATing.)
I think VM and Tivoli make the job of restoring our Linux environment
easy, but I'm sure that the same could be pulled off using native Linux
backups. One day I'm going to test the validity of the Tivoli Backup Set
concept, wherein you can supposedly restore an image directly from tape --
sans TSM server.
Matt Lashley
Systems Programmer
State of Idaho
James Melin
<[EMAIL PROTECTED] To:
[EMAIL PROTECTED]
epin.mn.us> cc:
Sent by: Linux on Subject: OS/390, Linux and
Disaster Recovery
390 Port
<[EMAIL PROTECTED]
ST.EDU>
12/17/2002 08:25
AM
Please respond to
Linux on 390 Port
What, if anything, have people done for disaster recover with Linux? How
do you restore Linux volumes at a disaster site? How evil is it?
